Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/273398616

MorphologicalevolutionoftheSouthPassageintheChangjiang(YangtzeRiver)estuary,China

ArticleinQuaternaryInternational·March2015

ImpactFactor:2.06·DOI:10.1016/j.quaint.2015.01.045

CITATIONS

3

READS

59

3authors:

ZhijunDai

EastChinaNormalUniversity

21PUBLICATIONS66CITATIONS

SEEPROFILE

JamesTLiu

NationalSunYat-senUniversity

106PUBLICATIONS1,219CITATIONS

SEEPROFILE

WenWei

EastChinaNormalUniversity

8PUBLICATIONS49CITATIONS

SEEPROFILE

Allin-textreferencesunderlinedinbluearelinkedtopublicationsonResearchGate,

lettingyouaccessandreadthemimmediately.

Availablefrom:JamesTLiu

Retrievedon:02June2016

lable at ScienceDirect

Quaternary International xxx (2015) 1e13

Contents lists avai

Quaternary International

journal homepage: www.elsevier .com/locate/quaint

Morphological evolution of the South Passage in the Changjiang(Yangtze River) estuary, China

Zhijun Dai a, *, James T. Liu b, Wei Wen a

a State Key Lab of Estuarine & Coastal Research, East China Normal University, Shanghai 20006, Chinab Department of Oceanography, National Sun Yat-sen, Kaohsiung 80424, Taiwan

a r t i c l e i n f o

Article history:Available online xxx

Keywords:Estuarine morphologyBranching estuaryShoal evolutionRiver runoffRiver sediment loadEOF analysis

* Corresponding author.E-mail address: zjdai@sklec.ecnu.edu.cn (Z. Dai).

http://dx.doi.org/10.1016/j.quaint.2015.01.0451040-6182/© 2015 Elsevier Ltd and INQUA. All rights

Please cite this article in press as: Dai, Z., et aQuaternary International (2015), http://dx.d

a b s t r a c t

Estuarine morphology of the world's major large rivers have experienced great changes due to intensiveanthropogenic activities and natural forcings, especially in the Changjiang Estuary located at the end ofthe longest river in Asia. This study focuses on morphological changes of the South Passage (SP), which isone of the 4 channels in the branching estuary at the mouth of the Changjiang (Yangtze River), China. Amultivariate analysis technique of Empirical Orthogonal/Eigen Function (EOF) method was used toexamine the major modes of change in the long-term (over 26 years) water-depth data. The results showthat the morphological changes at the SP could be divided into two stages: between 1987 and 1997, theSP had a single stable channel with closure of a cross-channel. Between 1997 and 2012, SP displayedsoutheastward elongation of a spit into the main channel, and westward shoal incision by a cross-channel. The opening of the SP developed a two-channel morphology, which stabilized and showedinfilling during 2003e2012. The average deposition rate was 10 cm/yr. In the past 30 years, the mostdominant morphological changes of SP included the deposition around the upstream opening of thechannel. The second most important pattern of morphological change was related to the downstreamelongation, retreat, and lateral migration of the spit of the Jiangyanan Shoal, which resulted in the two-channel configuration of the SP. Additionally, these morphological changes were not triggered by thedecline of the distal sediment source from the upstream, but due to the input of proximal sources of theshoal at the upstream opening of the SP and spill-over sediment from the North Passage via a short-cut-channel.

© 2015 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction

Estuaries are located in the narrow zone of the land-seaboundary, which are valuable areas of both local and globalimportance. The morphology of estuaries is closely linked to estu-arine assets such as natural location to ports (Van der Wegen andRoelvink, 2008). The morphological formation and evolution ofestuaries is relatively gradual and has a typical decadal timescale.Small-scale sea-level fluctuations may trigger drastic changes(Dyer, 1997). However, in the last century the gradual developmentof many estuaries and adjacent area has been interrupted by hu-man activities (Lane, 2004; Kao and Milliman, 2008; Evans, 2012;Gong et al., 2012; Li et al., 2012). Such practices include forexample, damming of the river, construction of dikes and groins in

reserved.

l., Morphological evolution ooi.org/10.1016/j.quaint.2015.0

the river mouth, and land reclamation (Chu et al., 2009; Syvitskiet al., 2009; Dai and Lu, 2010; Evans, 2012).

The reduction of river load due to damming is the cause for risksof erosion and inundation inmany estuaries and deltas in theworld(Walling and Fang, 2003; Syvitski et al., 2005). A well-knownexample is the Nile delta. Because of the Aswan Dam, the erosionrate for the area around Rosetta Promontory was 10 � 106 m3/y(Inman and Jenkins, 1984). Because of the construction of the damat Adana, the nearby Seyhan delta in Turkey has the erosion rate of1 � 106 m3/y Gagliano et al. (1981) estimated the land loss aroundthe mouth of the Mississippi was 102.1 km2 in 1980, which washigher than those of 1987, 1946, and 1913 at 72.8, 40.9, and17.4 km2/y, respectively. Compared to the previous century, theEbro River in Spain lost 99% of its load due to the constructions ofdams and reservoirs (Gull�en and Palanques, 1997). The Danube lost35% of its load due to dam construction that led to the erosion at theriver mouth (Milliman and Farnsworth, 2011). Since 1941, thesediment load in the Colorado River declined by almost 100%,

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Fig. 1. Maps showing: a. the location of the study area in relation to the China; b. themorphological units at the mouth of the Changjiang based on the 2006 navigationalchart; c. the location of the SP and the study area is marked by dashed lines.

Z. Dai et al. / Quaternary International xxx (2015) 1e132

which caused irreversible damages to its delta (Pitt, 2001). Becauseof the sediment load reduction, the Niger delta suffered a recessionrate of 10m/y (Ibe,1996). All of the examples above point to the linkbetween reduction in the river sediment load and the erosion in theestuary, and shoreline recession of the delta at the river mouth.

However, due to the various physiographic characteristics of theriver catchment (such as the nature of topography and river bed,spatial scale of the catchment, and degree of human disturbance),the change of suspended sediment load in the upper reaches of ariver might not be directly reflected by the load in the lower rea-ches (Brizga and Finlayson, 1994; Phillips et al., 2004). For example,no dam-related changes in alluvial sedimentation are noticeable inthe lower river reaches of southeast Texas, despite large reservoirscontrol 75e95% of the drainage area, and retention of massiveamounts of water behind dams (Phillips, 2003; Phillips et al., 2004).Channel dredging and jetty construction caused theMersey Estuarylost 0.1% of its volume in the past 150 years (Lane, 2004).Conversely, because of the retaining walls, a large amount ofsediment has accumulated in the middle of the Lune Estuary(Spearman et al., 1998). Furthermore, morphological changes of ariver mouth and its adjacent coast could also be caused by landreclamation, dredging of navigational channels, relative sea-levelrise, and littoral drift (Phillips and Slattery, 2006; Van Landeghemet al., 2009; Evans, 2012). Under multiple influences of thechanges in the river runoff and sediment load, changes in the localenvironment, engineering practices, and changes in the estuarinehydrodynamics, the morphological evolution of an estuarine iscomplex, and needs a comprehensive approach in the analysis.

The estuary of the Changjiang is located at the end of the longestriver in China. The width of the river mouth is almost 100 kmwide(Fig. 1). At the river mouth, the mean tidal range is 2.67 m, and themean flow rate is 1 m/s (Dai et al., 2013a). Three islands,Chongming, Hengsha, and Jiuduan Shoal divide the mouth into 4channels of North Branch, North Channel, North, and South Pas-sages (Fig. 1b). In the past century, the river mouth has been pro-grading seaward (Chen et al., 1985). Because of extensive landreclamation, the North Branch has become a tidally dominatedestuary (Yun, 2004). Due to the engineering project to maintain adeep navigational channel, the North Passage has become an arti-ficially controlled estuarine channel (Dai et al., 2013a) (Fig. 1). Theengineering project started in January 1998 (Fan and Gao, 2009),which consisted of four parts: the river diversion, construction ofnorth and south jetties along the channel, and dredging of thethalweg. The first stage of the project completed in June 2001,which achieved the diversion of the river flow, partial completionof the jetties, and the navigational channel reached 8.5 m depth.The second stage began in May 2002 and lasted until March 2005,inwhich the jetties with the attached groins along the length of theN. Passage were completed and the navigational channel reached10 m depth. The last stage commenced in September 2006 and wascompleted in March 2010, in which the dredging was the mainfocus and the navigational channel reached 12.5 m depth. Upon thecompletion of the channel improvement and maintenance project,the deep-water navigation channel was 92.2 km long, 350e400 mwide, having the thalweg depth of 12.5 m (Fan and Gao, 2009; Fanet al., 2012; Dai et al., 2013a) (Fig. 1).

Theworld's largest dam, the Three Gorges Dam (TGD), located inthe upstream of the Changjiang, became operational in 2003. Themeasured suspended sediment discharge (SSD) at Datong (theupper limit of the tidal river) dropped sharply to 2.06 � 108 and1.47 � 108 in 2003 and 2004, respectively, in comparison to over4.2� 108 t/y before the 1970s (BCRS, 2010, 2011). In 2006, and 2011,the sediment delivered to the sea declined further to be 0.84 � 108,0.72 � 108 t/y (BCRS, 2010, 2011). Whether the drastic reduction ofsediment load is going to trigger erosion or the progradation of the

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

Changjiang estuary delta will continue has been an internationaldispute (Liu et al., 2007; Yang et al., 2007, 2011; Syvitski et al., 2009;Wang et al., 2009; Chen et al., 2010; Dai et al., 2013b; Wang et al.,2013). Based on local cross-sectional changes in the estuary, Yanget al. (2007) found that the reduction of the river sediment loadof the Changjiang is the reason for the erosion of estuarine wet-lands. However, from the analysis of the morphological changes ofthe shipping channel along the North Passage, Dai et al. (2013a)found deposition in the peripheral groin fields along the shipping

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Table 2Hydrological gauging stations of the South Passage

Stations Position Surveyed periods

NC1 121.76� E, 31.27� N 8.2002, 8.2005, 8.2007, 8.2010NC2 121.85� E, 31.19� N 8.2002, 8.2005, 8.2007, 8.2010

Z. Dai et al. / Quaternary International xxx (2015) 1e13 3

channel and deepening of the shipping channel, which is primarilydue to the channel improvement and maintenance engineering.Variability of the river-mouth shoal (seaward migration and sizereduction), which is caused by the declining sediment discharge ofthe Changjiang due to the TGD, and the enhancement of the ebbflow as the result of dredging (Dai et al., 2013a). Furthermore, Jianget al. (2012) suggest that since 1999 the channel maintenance en-gineering practices have on one hand limited the seaward pro-gradation, but on the other hand, induced vertical accretion of theJiuduan Shoal. The combined factors of upstream sediment loadreduction and local channel maintenance engineering practices areno doubt making the morphological processes at the mouth ofChangjiang more complex.

Since both the North Branch and North Passage have beensubject to human alteration (such as local reclamation along theNorth Branch, channel dredging of the North Passage) (Yun, 2004),it is difficult to distinguish natural from the anthropogenic causes ofmorphological development. Although the North Channel has theleast human interference, it is not the major conduit for the exportof riverine sediment, and it has been the least studied. However, theSouth Passage (SP), being themajor conduit for water and sedimentof the Changjiang (Milliman et al., 1985), has also been affected bydirect human activities to a limited extent. Since 1987, this channelhas been monitored for navigational purposes with uninterruptedyearly nautical topographic records. Thereafter, the objective of thispaper is to diagnose the decadal morphological changes of thischannel to determine the whether the decline of distal river sedi-ment source or the proximal local influences would affect thedeveloptment of the SP. The outcome will be valuable to assess thepotential response of part or thewhole of the Changjiang Estuary tothe upstream reduction of sediment supply, and the impact of thenatural and man-made factors near and far. It could be applied toriver estuaries with similar distributary mouths worldwide that arealso under similar natural and anthropogenic influences.

2. Material and methods

Digitized bathymetry-derived DEM has become a useful tool tostudy morphological changes of estuaries (Thomas et al., 2002; VanderWal et al., 2002; Lane, 2004; Blott et al., 2006; Jaffe et al., 2007).This study collected surveyed bathymetric data of the SP since 1987.The 1987e2000 data came from published nautical charts by theMaritime Survey Bureau of Shanghai, Ministry of Communicationsof China. The data for 2001e2012 came from the nautical chartspublished by the Changjiang Estuary Waterway AdministrationBureau (CJWAB), Ministry of Transportation of China (Table 1).

Table 1Observed water depth data in the South Passage, Changjiang.

Surveyed date Map title Scale Data sources

1987, 1988, 1989, 1990, 1990, 1991, 1992, 1993, 1994, 1995, 1996,1997, 1998, 1999, 2000

Nancao Fairway 1:50000 MSBSMCRa

2001, 2002, 2003, 2004, 2005 Beicao and Nancao Fairway, Changjiang Estuary 1:25000 CJWABb

2006, 2007, 2008, 2009, 2010, 2011, 2012 Beicao and Nancao Fairway, Changjiang Estuary 1:10000 CJWAB

a MSBSMCR: Maritime Survey Bureau of Shanghai, Ministry of Communications of China.b CJWAB: Changjiang Estuary Waterway Administration Bureau (CJWAB), Ministry of Transportation of China.

All the bathymetric surveys were conducted in August orSeptember each year based on the theoretical low-water datum.Dual-frequency echo sounders and GPS positioning were used. Thevertical error was limited to be within 0.1 m. Survey density con-sisted of 8e20 bathymetric data points per km2. Maps of the

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

Changjiang Estuary and SP with the same precision have beenapplied to studies on estuarine erosion and deposition (Yang et al.,2011), evolution of the river trough (Jiang et al., 2012; Wang et al.,2013; Dai et al., 2013a), and the morphological changes of the es-tuary and submerged delta (Dai et al., 2014). Tthe morphologicalchanges in the current study exceed 0.1 m. Accordingly, the sur-veyed the digital data of the SP is reliable to use for the study of themorphological changes of the SP.

All charts were georeferenced using nine fixed benchmarks, ofwhich the National Grid coordinates were known. The depth valueswere converted to elevation relative to Beijing-1954 Datum. Afterobtaining the charts, we digitized them by using ArcGIS 9.3 as theplatform for input, output, and comparison. The Kriging schemewas used to interpret the digitized data into grid having resolutionof 250 � 250 m.

In the meantime, we collected the river flow data in the studyarea (source: CJWAB). The tidal flow was measured at S1 and S2 inthe SP in the flood season of August in 2002, 2005, 2007, and 2010,respectively (Table 2). The measurements at the two locations weresynchronized and took place during spring tide and lasted over25 h.

This study also collected the river runoff and suspended sedi-ment load at the Datong station located at the landward limit of thetidal river (Fig. 1, Hydrological Committee of Changjiang, www.cjw.com.cn) (BCRS, 2010, 2011).

At each hydrographic station, the coefficient of flow dominance(defined as tidal flow rate during the ebb divided by the sum of tidalflow rates of the ebb and flood) was calculated (Simmons, 1955) asfollows:

A ¼ Qe

Qe þ Qf� 100% (1)

in which A is coefficient of dominance flow, Qe is cross-sectionallyaveraged ebb flow rate, Qf is cross-sectionally averaged flood flowrate. When A> 50%, the flow is ebb-dominant. Conversely, it isflood dominant.

To quantify morphological changes, the �5-m isobath wasextracted through the ArcGIS platform. From ArcGIS the maximumand mean water depths of each chart were obtained. To quantifyerosion/deposition, the volume changes were calculated at 3e4year increments between 1987 and 2012, which included 26 yearly

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Z. Dai et al. / Quaternary International xxx (2015) 1e134

charts. The spatial and temporal correlations (standardizedcovariance) of the water depths in the 26-year period were thenanalyzed by using the multivariate analysis technique EOF(Empirical Orthogonal/Eigen Function) analysis.

EOF is a classical statistical technique, which was used toseparate orthogonal modes contained in the data sets (Liu et al.,2000). The advantage of using EOF analysis is that a set of inter-correlated variables can be decomposed into a set of statisticallyindependent modes, which bear physical meanings. Thus, the EOFtechnique has been widely applied to problems in meteorology,oceanography, geology, and sedimentology (Liu et al., 2000; Emeryand Thomson, 2001; Liu and Lin, 2004; Lane, 2004; Dai et al.,2013a). Here, all the standardized bathymetric variance sets canbe deviated to form a single matrix X

X ¼ ðX1…XnÞ

where each Xn represents a single grouped observation of the mvariables and X is an m � n data matrix. m is the number of spatialpoints, and n is the observed time points.

Mathematically, the EOFs are the eigenfunctions of the covari-ance matrix S, which is given by S ¼ XX0. Based on singular valuedecomposition of thematrix S, the eigenvalues l¼ (l1, l2,…, lm) of Sand spatial patterns (V) can be obtained. Thereafter, the time seriesdescribing the loadings (T) can be obtained by T¼ V0X (Wallace andDickinson, 1972).

Subsequently, eigenvalues l ¼ (l1, l2,…,lm) can be ranked ac-cording to the amount of data (covariance) they explained, i.e. thevariance contribution of each eigenvalue RK, and the cumulativecontribution for the eigenvalues G can be considered as:

Rk ¼lkPmi¼1 li

½k ¼ 1;2;…pðp<mÞ�

G ¼Pp

i¼1 liPmi¼1 li

ðp<mÞ

Generally, the better correlated the data are, the more covari-ance can be accounted for by the first few eigenmodes (Liu et al.,2000; Emery and Thomson, 2001). The remnant eigenfunctionsare neglected due to the effect of noise contained in this set(Wallace and Dickinson, 1972).

3. Results

3.1. Bathymetric changes along the SP

Jiangyanan Shoal is located at the upstream opening of the SP.Prior to 1997, Jiangyanan Shoal and Jiuduan Shoal were

Fig. 2. Nautical charts of

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

disconnected, as the downstream part of the South Channel(Fig. 2a). Currently, Jiuduan Shoal is located on the north side of theSP, and Nanhui Shoal is on the south side of the SP (Figs. 1c and 2b).Based on the changes of the location of the �5-m isobath (Figs. 1cand 3), themorphological changes at the SP can be characterized by2 stages. In the first stage of 1987e1997 there was a single channelat the SP, which is clearly shown in the nautical chart of 1990(Fig. 2a). In the early period of this stage the �5-m isobaths ofJiangyanan Shoal and Jiuduan Shoal were not connected. There wasa short channel about 1 km wide separating the two sand bodies(Fig. 3a), allowing exchange between the North and South Passages.Along the south side of the SP there was a ‘spur’ on the �5 m iso-bath pointing to the spit of the Jiangyanan Shoal (Fig. 3a). In thefollowing years of 1990e1994, the �5-m isobaths of Jiangyananand Jiuduan Shoals connected. The north side of the short-cut-channel was filled in. In the meantime, the spit at the southernend of the Jiangyana Shoal migrated northward by 1260 m (Fig. 3b)but the thalweg of the SP was located toward the north side of thechannel along Jiuduan Shoal. In 1994e1997, the short-cut-channelwas completed closed as shown by the �5-m isobath. The spit ofJiangyanan Shoal was eroded by 1060 m (Fig. 3c). The JiangyananShoal and Jiuduan Shoal were merged so that the bank of SPextended to include the Jiangyanan Shoal. Thereafter, between1987 and 1997, with closure of the short channel, the SP had asingle stable channel. As a result, the SP had a relatively straightthalweg. The short-cut-channel became a wide flood cove.

In the second stage of morphological changes in 1997e2012, thespit of the Jiangyanan Shoal elongated downstream, making theopening of the SP into a dual channel system (Fig. 3d and e). In theperiod of 1997e2000, the erosional pattern of the spit turned toaccretionary and the spit extended downstream by 6430 m. In thesame time, the �5-m isobath of the southern bank of the SPextended northward, to form a narrow gap of 800 m between thespit of the Jiangyanan Shoal and the southern bank of the SP(Fig. 3d). The flood cove between the two shoals narrowed and itsnorthern end moved downstream by 1670 m in response to theelongation of the spit (Fig. 3d). In 2000e2003, the spit of Jian-gyanan Shoal continued to accrete into the relative stable channelof SP by 2380 m. However, the entire spit shifted northward by1300 m together with the flood cove on its north side (Fig. 3e).

As the spit of the Jiangyanan Shoal grew in the southwesterndirection, the SP became a shoaletrough complex, which is clearlyreflected in the 2006 nautical chart (Fig. 2b). The spit increments inthe periods of 2003e06, 06e09, and 09e12 were 1200, 4100, and3500 m, respectively. The flood tidal cove on the north side of thespit became narrower (near 800 m in 2006 and less than 500 m in2012). Comparing to 2009, the position of the flood tidal covemoved NW by 6800 m (Fig. 3feh). Cross-sectional wide, due to the

a. 1990, and b. 2006.

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Fig. 3. Geomorphological changes of the SP and the two shoals as expressed by the �5-m isobath in 3-y increments.

Z. Dai et al. / Quaternary International xxx (2015) 1e13 5

Please cite this article in press as: Dai, Z., et al., Morphological evolution of the South Passage in the Changjiang (Yangtze River) estuary, China,Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.01.045

Fig. 4. Water depth changes of the SP in 1987e2012: a. yearly mean depths; b. yearly maximum depths.

Z. Dai et al. / Quaternary International xxx (2015) 1e136

downstream accretion of the spit in the upper part of SP, the lowerpart of the SP deepened and the thalweg remained closer to theJiuduan Shoal. Thus, the present-day SP formed a complexmorphology.

3.2. The bathymetry and associated volume change at the SP

The yearly average water depth of the study area in the 30-yearperiod between 1987 and 2012 fluctuated between 5.1 and 6.7 m,by over 1 m, but also showed a decreasing trend (Fig. 4a). A suddendrop over 1 m occurred in 1999. During the same period, themaximum water depths also showed a decreasing trend (Fig. 4b),which had a sudden decrease from 9m in 1998 to 7 m in 2000, alsoreflected in the drop of the yearly mean depth (Fig. 4a).

The volume capacity of SP has been gradually decreasing from0.69 � 109 m3 at 1987 to 0.59 � 109 m3 at 1994, a reduction by 14%(Fig. 5), implying infilling of the SP. In 1994e1996, the capacityincreased by 0.04 � 109 m3. After 1998, the capacity continued todecrease. In 2001e2006, the capacity increased slightly. In2006e2008, the capacity dropped drastically by 0.06 � 109 m3, andthen gradually gained back slightly. Overall, between 1987 and2012, the volume capacity of SP lost about 13% (Fig. 5). Comparingto the 0.1% volume decrease of the Mersey Estuary in the past150 yr (Lane, 2004), the decrease at SP is considerable.

Fig. 5. The volume capacity change below the 0-m isobath of the SP in 1987e2012.

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

3.3. Deposition/erosion of the SP

Visual comparisons of the depth changes in 3e4 y increments(Fig. 6aeh, 7) indicate that the stages of erosion/deposition patternsof the SP were identical to the 2 stages described earlier. In the firststage (1987e1997) the SP was more or less in erosion/depositionequilibrium, having slight tendency of deposition. Because of thejoining of the spit of Jiangyanan Shoal and the Jiuduan Shoal, theupstream part of the SP turned from erosional (by 1 m) to depo-sitional (by 1 m) (Fig. 6b and c). The overall deposition was about0.2 m.

In the second stage (1997e2012), SP was firstly in the state ofadjustments during 1997e2003 (Figs. 5d, 6e and 7). This is stronglyreflected by the deposition around the upstream opening of the SP.Compared to the previous period of 1994e1997 (Fig. 6c), in theperiod of 1997e2000 the entire SP experienced slight net erosion.The erosional rate was over 5 cm/y (Fig. 7). In 2000e2003, the totalaccumulation around the upstream opening of the SP was over 4 m.In contrast, around the seaward opening near Jiuduan Shoal, thetotal accumulation was over 3 m in 1994e1997. In 1997e2000, theerosional rate was over 6 cm/y (Fig. 7). But in 2000e2003, a slightaccumulation rate of about 10 cm/y occurred (Fig. 7). From 1997 to2003, the net change at SP was from an erosion rate of 5 cm/y to anaccumulation rate of about 10 cm/y (Fig. 7).

In the subsequent years during 2003e2012, the SP was in a stateof strong deposition that had a net accumulation rate over 10 cm/y(Fig. 7). Within this trend, localized strong erosion and depositionalso occurred (Fig. 6feh). These areas included the continuedaccumulation at the upstream opening of the SP by 1e2 m. Thesouthern end of the SP on the other hand, was in the state oferosion by about 1e2 m. Deposition also occurred in the centersection of the SP, on average about 1 m. Since 2003e2009, thedeposition area in the center-channel gradually moved upstream(Fig. 6f and g). In 2009e2012, the entire SP was in a general state ofaccumulation (average rate of 10 cm/y) except for a narrow zone oferosion over 1 m along the center section of Jiuduan Shoal (Figs. 6hand 7). Thereafter, between 1997 and 2003, the SP went throughanother period of erosion/deposition adjustment during which theopening of the SP developed a two-channel morphology (Fig. 7),which stabilized and showed infilling during 2003e2012 (Fig. 7). Inthis entire period (1997e2012) the changes included eastward spit

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Fig. 6. Erosional/depositional patterns of the SP (negative means erosion, positive means deposition) in different time periods.

Fig. 7. The net erosion/deposition values of the SP derived from Fig. 6.

Z. Dai et al. / Quaternary International xxx (2015) 1e13 7

Please cite this article in press as: Dai, Z., et al., Morphological evolution of the South Passage in the Changjiang (Yangtze River) estuary, China,Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.01.045

Z. Dai et al. / Quaternary International xxx (2015) 1e138

formation and westward shoal incision of the short channel, east-ward spit formation and westward shoal incision of the cross-channel.

3.4. EOF analysis of the bathymetric changes

The results of EOF analysis on the water depth data from 1987 to2012 show that the first two eigenmodes can explain over 75% ofthe standardized covariance of the data (Fig. 8). As all higher modesexplain less than 10% of the data, only the first two modes arepresented and discussed.

The first eigenmode explains 59% of the water depth co-variability, which is the most important form of morphologicalchanges. The contour plot of the eigenvectors of this mode showsthe spatial co-variability in the SP including the accretion at thecenter of the northern opening of the channel (the area of highestpositive values) due to the sediment load input from upstream andalong the south side of the Jiuduan Shoal (Fig. 8a). Also, along thenorthern edge of the SP bordering the Jiuduansha there is anelongate area having values of 0e0.05. These are the areas ofaggradation whose temporal pattern shows increasing trend asindicated by the eigenweighting curve of this mode (Fig. 8b). Thismode also shows the formation of the thalweg in the center of thechannel. The temporal pattern of this mode is expressed by theeigenweighting curve of this mode (Fig. 8b), which shows anincreasing secular trend, which resembles the inverse trends of thedecreasingmeanwater depth, maximumdepths (Fig. 4a and b), andthe decreasing volume capacity (Fig. 5). Therefore, this mode de-scribes the increasing spatial gradients in sediment transport intothe SP whose most significant effect is shown by the major depo-sition in the upstream opening of SP depicted by the eigenvectors(Fig. 8a).

The second mode explains the 18% of the water depth co-variability. The eigenvectors of this mode indicate that themorphological changes related to the growth, decay and move-ments of the spit of the Jiangyanan Shoal, which is the second mostimportant factor contributing to the water depth changes in the SP(Fig. 8c). The ‘W’ shaped temporal pattern suggests a somewhatrecurring pattern, if not cyclical. From 1987 to 1995, the descendingtrend in the eigenweighting curve reflects the retreat of the spit of

Fig. 8. The EOF analysis result of the time-space related water-depth changes Showing: a. ththe 2nd mode eigenvectors; and d. the 2nd mode eigenweighting curve.

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

the Jiangyanan Shoal (Figs. 8d and 3). On the other hand, the rapiddownstream elongation of the spit caused the ascending trendbetween 1995 and 1998 (Figs. 8d and 3). Between 1999 and 2002the spit retracted and became shorter. In the meantime, it migratedtowards the Jiuduan Shoal, making the tidal cove narrower (Figs. 8dand 3). Beginning in 2003, the spit began to lengthen again,accompanied by the erosion and the head-ward retrogression ofthe cove, making it resemble a narrow and long channel (Fig. 3f).This development is reflected by the ascending part of the eigen-weighting curve after 2003 (Fig. 8d).

4. Discussion

4.1. The changing sediment load of Changjiang

The average water discharge of Changjiang exceeds900 � 109 m3/y. The runoff of the Changjiang did not shownoticeable change in the period between 1954 and 2011 (Fig. 9a).The mean sediment load of the Changjiang is 420 � 106 m3. In theperiod of 1954e2011 therewas a decreasing trend, whichwasmorepronounced in the period of 1987e2011 (Fig. 9b). In the case of theNP (Fig. 1a), the accumulation is on the order of hundreds of milliontons despite the sediment load reduction (Jiang et al., 2012; Daiet al., 2013a). This is largely attributed to the engineering works(Dai et al., 2013a). The regions shallower than the �5-m isobath onthe submerged delta of the Changjiang was still in the state of ac-cretion (Yang et al., 2011). Controversy continues regardingwhether or not the shores around themouth of Changjangwill startto erode (Chen et al., 2010; Yang et al., 2011; Dai et al., 2012, 2013a;Wang et al., 2013).

Sediment supplied by the Changjiang determines the seawardprogradation of its river mouth (Chen et al., 1985). Since the 12thcentury, the shoals on the south bank of the Changjiang haveprograded seaward at the pace of 1 km per 40 y. In the last halfcentury, the rate accelerated to 1 km per 23 y (Chen et al., 1985).Since the SP is the major seaward conduit of the riverine sediment(Milliman et al., 1985), it should show responses to the loadreduction. The relations among the maximum water depth, meandepth, and the whole capacity below 0 m isobath of the SP and thesediment load of Changjiang into the estuary in the period

e plot of the 1st mode eigenvectors; b. the 1st mode eigenweighting curve; c. the plot of

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Fig. 9. Yearly records of discharge (a), and suspended sediment load (b) at the Datong st. The black lines are the linear regression results for the period of 1954e2011 and the redlines are the linear regression results for the period of 1987e2011, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the webversion of this article.)

Fig. 10. Yearly mean sediment load at Datong st. verses (a) yearly max. water depth; (b) yearly mean water depth; and (c) volume capacity below the 0-m isobath of the SP.

Z. Dai et al. / Quaternary International xxx (2015) 1e13 9

1987e2011 show statistically significant (p < 0.001, except for themaximum water depth) positive correlations (Fig. 10aec). Thesepatterns indicate that the decrease in mean and maximum waterdepths and the volume capacity of the SP are associated with thereduction of the sediment supply, which is counter-intuitive. Thevalue of mean grain-size of the Changjiang sediment was0.0017 mm (1976e2000) and that of the medium diameter in 2012was of 0.011 mm. The fining trend corresponds with the depositiontrend in the SP, attesting to the deposition in the SP. On the otherhand, reduction of sediment supply should have induced erosion,which in turn would have caused deepening of the SP and an in-crease of the volume capacity. However, the above relations pointto factors other than these direct distal effects of the reduction ofsediment supply upstream of the Changjiang Estuary. This isconsistent with the recent discovery that the despite the reductionof the distal sediment source, the submerged delta of the Chang-jiang still showed aggradation (Dai et al., 2014) due to proximal andlocal factors.

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

4.2. Local influences

Engineering works in the NP have affected the Jiangyanan andJiuduan Shoals. The construction of the south jetty and dike systemin the NP stabilized the northern rim of these two shoals. From thesecond stage to the end of the third stage of the navigation channelimprovement and maintenance project the flow dominance coef-ficient measured at the two stations in the SP first showeddecreasing and then increasing trends (Fig. 11). The minimum ebbdominance was 60%, which reached 70% during the flood season of2010. These trends indicated the engineering practices of theproject strengthened the ebbing flow, which enhanced the chan-nel's ability to transport sediment seaward (Fig. 11). Furthermore,the engineering structures blocked the sediment exchanges be-tween the NP and SP (Yang et al., 1998; Li et al., 2006). This is re-flected by the disappearing of the short-cut-channel before theearly 1990s (Fig. 3aec). In 1997e2006, during slack transition fromflood to ebb, the suspended sediment in the NP was still able to

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Fig. 11. The measured flow dominance coefficients at two hydrographic stations atdifferent times.

Z. Dai et al. / Quaternary International xxx (2015) 1e1310

move over the submerged dike and enter the SP (Liu et al., 2010).Since the peripheral groin field of the NP trapped sediment, thiscould be a local source for SP. The depositional pattern along theJiuduan Shoal illustrated by the 1st mode eigenvectors (Fig. 8a)

Fig. 12. Schematic plot of the land reclamation of th

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

could be the effect of this source. However, this sediment inputfrom NP did not fully fill in the tidal cove, which remained stable(Fig. 3deg). Rather, sediment could probably bypass the cove andenter the main channel of SP. In 2012, the cove almost cut Jian-gyanan Shoal in the form of head-ward erosion, which waspotentially caused by the cut-off of spill-over sediment from the NP.

Moreover, the SP has not been significantly affected by landreclamation, which was carried out on the tidal flat of Nahui on thesouth bank of the SP in 1994e2003 (Figs. 1b and 12) (Li et al., 2007).At present, the reclaimed area above the �2 m isobath exceeds 15,000 hm2 (Li et al., 2010). Although the geomorphology of SP in1987e2012 has shown appreciable changes (Fig. 3) in the riverchannel, but the positions of the �5-m isobath on both sides of theSP remained unchanged and relatively stable. This suggests thatreclamation-engineering projects at the seaward end of the SP hadlittle influence on the SP morphology, although there may havebeen an impact on shallower areas.

4.3. Spit morphodynamics of the Jiangyanan Shoal

The EOF analysis points out that the most important mode ofmorphological change is characterized by the deposition around

e Nanhui Shoal (Modified from Li et al., 2007).

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Z. Dai et al. / Quaternary International xxx (2015) 1e13 11

the northern opening of the SP. To further illustrate a possibleproximal sediment source, the development and evolution of theJiangyanan Shoal is expressed by the 0-m isobath at discrete timepoints from 1994 to 2009 (Fig. 13).

In 1994, only a few small-scattered sand bodies were present atthe general location of the present Jiangyanan Shoal following itsorientation (Fig. 13a). In 1997, the size of the sand bodies increasedand they continued to develop along the orientation of the Jian-gyana Shoal (Fig. 13b). In 1999, the small bodies merged into onelarge body forming the Jiangyanan Shoal (Fig. 13c). In the mean-time, the southern limit of the shoal moved 990 m downstreamcompared to the 1997 location. In 2002, Jiangyanan Shoal moveddownstream by another 3000m, whose southern tip was separatedfrom the main body to form a small sand body (Fig. 13d). In 2006the Jiangyanan Shoal moved farther downstream by 1990 m as onebody (Fig. 13e). In 2009 the Jiangyanan Shoal moved 1700 m andthe elongated spit was well formed (Figs. 3g and 13f). These historicchanges suggest the development and downstream movements ofthe Jiangyanan Shoal also contributed to the infilling of the SP.

The downstream movement was probably associated with thelateral expansion of the Jiangyanan Shoal into the subaerial SPchannel, which also can be reflected by the 0-m isobaths (Fig. 13).The downstream movement of the shoal was facilitated by the bedmaterial along the SP. In the Changjiang Estuary, shoal movementsare a major form of bedload sediment transport that amounts to350 � 106 t/y (Li, 1993; Li et al., 2010). This amount is equivalent tothe amount of sediment delivered by the Changjiang to the sea,which on a large degree, results in the shoaling of estuary channels(Li, 1993; Li et al., 2010). Generally these shoals comprise of finesand having the medium diameter of 0.16 mm (Zhou, 1983). Understrong ebbing flow (average velocity: 1e1.5 m/s) the sedimenttransport in shallow part of the Changjiang Estuary is in the form ofsand wave movement, whose migration rate could reach300e2000 m/y in the flood season (Zhou, 1983; Li, 1993; Gonget al., 2003). Li (1993) described the mechanism of the sand wavemigration. This implies that the sediment on the Jiangyanan Shoalcould move into SP via sand waves. Thereafter, the temporary trendof the 1st eigenmode thus, could also reflect the enlargement of the

Fig. 13. Time changes of the 0-m isobath of the Jiangyanan Shoal (0-m isobath: do

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

Jiangyanan Shoal, which is reflected by the deposition next to theshoal.

The second most important pattern of morphological changeswas caused by the slow accretion, recession, and lateral migrationof the spit of the Jiangyanan Shoal (Fig. 3aeh). In 1997e2000, thespit extended into the SP channel by 6430 m. The extension causedthe single-channel thalweg to become a double-trough channel(Fig. 3def). Moreover, the spit also displaced the thalweg and madeit shift laterally (Fig. 3). A critical time point in the spit-morphodynamics is 1998 when historically the second most se-vere flooding in Changjiang occurred (Fig. 9). The river dischargeexceeded 120 � 109 m3. The strong river flow and large amount ofsediment load probably triggered the spit of the Jiangyanan Shoalto extend downstream, causing the mean water depth to decreasenoticeably between 1998 and 2000 (Fig. 4). Once this spit was inplace, in the ensuing years, the spit-building process continuedwith fluctuations.

At present, the dikes along the south side of the navigationchannel stabilized the northern rim of the Jiangyanan Shoal. Jiu-duan Shoal is in the state of equilibrium (Jiang et al., 2012). As thereduction in the sediment load from the upstream continues, andthe strength of the river flow remains unchanged (Fig. 11), the spitin the SP might not continue to extend downstream. The SP mightrevert to a single trough channel and the seaward extension of theJiangyanan Shoal might slow down or stop, and the Jiuduan Shoalmight undergo erosion.

5. Conclusions

Having a branching estuary, the Changjiang has one of the mostcomplex river mouths in the world. In the face of declining sedi-ment supply and extensive engineering practices, different chan-nels of this branching estuary have displayed differentialmorphological processes as followed as:

1. Between 1987 and 2012, the average, maximum, and the vol-ume capacity below the 0-m isobath were all in a decreasingtrend. The volume capacity in the last 30 years has decreased by

tted line; the southern jetty of the navigational channel and dikes: dark line).

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Z. Dai et al. / Quaternary International xxx (2015) 1e1312

13%. The erosion/deposition patterns varied in time. In1987e1997 the SP was in a state of equilibrium. The net depthchange was 0.2 m. In 1997e2003, the SP was in a state oftransition, which was followed by a state of infilling by theamount of 10 cm/y between 2003 and 2012. The infilling is themost important mode of morphological changes indicated bythe EOF analysis result, which is mostly attributed to thedeposition in the northern opening of the SP.

2. There were two stages in the morphological processes at the SP.Between 1987 and 1997 the SP had a single channel with closureof the cross-channel. Between 1997 and 2012, the SP firstlytransitioned from a single channel to a two-channel configura-tion from 1997 to 2003. From 2003 to 2012, the SP remained atwo-channel branch. In this period SP showed eastward spitformation and elongation and westward shoal incision of thecross-channel. These two stages were controlled by thelengthening, retreat, and lateral migration of the spit of theJiangyanan Shoal, which is the second most important mode ofmorphological changes indicated by the EOF analysis results.

3. The secular trend in the sediment load decrease cannot explainthe opposing trends of water depth and volume capacitydecrease at the SP. The morphological changes in the SP weremainly induced by the development, growth, and migration ofthe Jiangyanan Shoal into the SP and the morphodynamics ofthe spit of the Jiangyanan Shoal. Furthermore, the engineeringproject north of the SP stabilized the river flow partition andconstricted the ebbing flow to be stronger as shown by the highcoefficient of ebb-flow dominance. This might help to transportsediment into SP from the upstream.

Acknowledgements

This study was supported by the National Science Foundation ofChina (41476076, 41076050) and New Century Excellent Talents inUniversity of China (NCET-12-0182). We are grateful to Dr. Mick vander Wegen and two anonymous reviewers for their constructivecomments and suggestions that improved the article.

References

Bulletin of China River Sediment (BCRS), 2010. Press of Ministry of Water Resourcesof the People's Republic of China. http://www.cjh.com.cn/.

Bulletin of China River Sediment (BCRS), 2011. Press of Ministry of Water Resourcesof the People's Republic of China. http://www.cjh.com.cn/.

Blott, S.J., Pye, K., van der Wal, D., Neal, A., 2006. Long-term morphological changeand its causes in the Mersey Estuary, NW England. Geomorphology 81,185e206.

Brizga, S.O., Finlayson, B.L., 1994. Interactions between upland catchment andlowland rivers: an applied Australian case study. Geomorphology 9, 189e201.

Chen, J.Y., Zhu, H.F., Dong, Y.F., Sun, J.M., 1985. Development of the Changjiang es-tuary and its submerged delta. Continental Shelf Research 4 (1/2), 47e56.

Chen, Z.Y., Wang, Z.H., Finlayson, B., Chen, J., Yin, D.W., 2010. Implications of flowcontrol by the Three Gorges Dam on sediment and channel dynamics of themiddle Yangtze (Changjiang) River, China. Geology 38 (11), 1043e1046.

Chu, Z.X., Zhai, S.K., Lu, X.X., Liu, J.P., Xu, J.X., Xu, K.H., 2009. A quantitativeassessment of human impacts on decrease in sediment flux frommajor Chineserivers entering the western Pacific Ocean. Geophysical Research Letters 36 (19).http://dx.doi.org/10.1029/2009GL039513.

Dai, S.B., Lu, X.X., 2010. Sediment deposition and erosion during the extreme floodevents in the middle and lower reaches of the Yangtze River. Quaternary In-ternational 226, 4e11.

Dai, Z.J., Liu, J.T., Xie, H.L., Shi, W.Y., 2012. Sedimentation in the outer Hangzhou Bay,China: the influence of Changjiang sediment load. Journal of Coastal Research.http://dx.doi.org/10.2112/JCOASTRES-D-12-00164.1.

Dai, Z.J., Liu, J.T., Fu, G., Xie, H.L., 2013a. A thirteen-year record of bathymetricchanges in the North Passage, Changjiang (Yangtze) estuary. Geomorphology187, 101e107.

Dai, Z.J., Chu, A., Li, W.H., Li, J.F., Wu, H.L., 2013b. Has suspended sediment con-centration near the mouth bar of the Yangtze (Changjiang) estuary beendeclining in recent years? Journal of Coastal Research 29 (4), 809e818.

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

Dai, Z.J., James, T.L., Wei, W., 2014. Detection of the Three Gorges Dam influence onthe Changjiang (Yangtze River) submerged delta. Scientific Reports 4, 6600.http://dx.doi.org/10.1038/srep06600.

Dyer, K.R., 1997. Estuaries: a Physical Introduction, second ed. John Wiley and Sons,Inc, p. 195.

Emery, W.J., Thomson, R.E., 2001. Data Analysis Methods in Physical Oceanography,second ed. Elsevier, Amsterdam, p. 638.

Evans, G., 2012. Deltas: the fertile dustbins of the continents. Proceedings of theGeologists’ Association 123, 397e418.

Fan, Q.J., Gao, M., 2009. Improvement project of deep-draft channel of YangtzeEstuary. Yangtze River 40 (8), 25e30 (in Chinese with English abstract).

Fan, Z.Y., Ge, J.Z., Ding, P.X., Pan, L.Z., 2012. Effect of the deep waterway project onthe salinities in the north passage of the Changjiang Estuary. Journal of EastChina Normal Univerisity (Natural Science) 4, 181e189 (in Chinese with Englishabstract).

Gagliano, S.M., Meyer-Arendt, K.J., Wicker, K.M., 1981. Land loss on the Mississippiriver deltaic plain. Gulf Coast Association Geological Society Transactions 31,295e300.

Gong, C.L., Yun, C.X., Yu, Z.Y., 2003. Bed-load movement in the south channel and itseffect on the south and north passages in the Yangtze Estuary. The Ocean En-gineering 21 (3), 62e67.

Gong, W.P., Wang, Y.P., Jia, J.J., 2012. The effect of interacting downstream brancheson saltwater intrusion in the Modaomen Estuary, China. Journal of Asian EarthSciences 45 (2), 223e238.

Gull�en, J., Palanques, A., 1997. A historical perspective of the morphological evo-lution in the lower Ebro River. Environmental Geology 30, 174e180.

Ibe, A.C., 1996. The Niger Delta and sea-level rise. In: Milliman, J.D., Haq, B.U. (Eds.),Sea-level Rise and Coastal Subsidence. Kluwer Academic Publishers,Netherlands, pp. 249e267.

Inman, D.L., Jenkins, S.A., 1984. The Nile littoral cell and man's impact on the coastalzone of the South East Mediterranean. In: 19th Coastal Engineering Conference.Proceedings American Society of Civil Engineers, Houston, Texas,pp. 1600e1617.

Jaffe, B.E., Smith, R.E., Foxgrover, A.C., 2007. Anthropogenic influence on sedimen-tation and intertidal mudflat change in San Pablo Bay, California: 1856e1983.Estuarine, Coastal and Shelf Science 73, 175e187.

Jiang, C.J., Li, J.F., de Swart, H.E., 2012. Effects of navigational works on morpho-logical changes in the bar area of the Yangtze Estuary. Geomorphology139e140, 205e219.

Kao, S.J., Milliman, J.D., 2008. Water and sediment discharges from small moun-tainous rivers, Taiwan: the roles of lithology, episodic events, and human ac-tivities. Journal of Geology 116 (5), 431e448.

Lane, A., 2004. Bathymetric evolution of the Mersey Estuary, UK, 106-1997: causesand effects. Estuarine, Coastal and Shelf Science 59, 249e263.

Li, J.F., 1993. The bedload movement in the Changjiang estuary. China Ocean En-gineering 7 (4), 441e450.

Li, J.F., Wan, X.N., Ying, M., Zuo, S.H., Fu, G., 2006. Study on the formation andevolution of Jiuduan shoal in the Changjiang estuary. Journal of SedimentResearch 6, 44e49 (in Chinese with English abstract).

Li, J.F., Dai, Z.J., Ying, M., Wu, R.R., Fu, G., Xu, H.G., 2007. Analysis on the developmentand evolution of tidal flats and reclamation of land resource along shore ofShanghai city. Journal of Natural Resources 22 (3), 361e371 (in Chinese withEnglish abstract).

Li, J.F., Dai, Z.J., Liu, X.C., Zhao, J.C., Feng, L.X., 2010. Researchon themovement ofwaterand suspended sediment and sedimentation in Nanhui spit of the Yangtze es-tuary before and after the cdonstruction of reclamation projects on the tidal flat.Journal of Sediment Research 3, 31e37 (in Chinese with English abstract).

Li, P., Yang, S.L., Milliman, J.D., Xu, K.H., Qin, W.H., Wu, C.S., Chen, Y.P., Shi, B.W.,2012. Spatial, temporal, and human-induced variations in suspended sedimentconcentration in the surface waters of the Yangtze Estuary and adjacent coastalareas. Estuaries and Coasts 35 (5), 1316e1327.

Liu, J.T., Lin, H.L., 2004. Sediment dynamics in a submarine canyon: a case of river-sea interaction. Marine Geology 207, 55e81.

Liu, J.T., Huang, J.S., Hsu, R.T., Chyan, J.M., 2000. The coastal depositional system of asmall mountainous river: a perspective from grain-size distributions. MarineGeology 165, 63e86.

Liu, J.P., Xu, K.H., Li, A.C., Milliman, J.D., Velozzi, D.M., Xiao, S.B., Yang, Z.S., 2007. Fluxand fate of Yangtze River sediment delivered to the East China Sea. Geo-morphology 85 (3e4), 208e224.

Liu, J., Zhao, D.T., Cheng, H.F., 2010. Recent morphological evolution of Jiuduan Shoaland its effect on navigation channel sedimentation of North Passage. Journal ofYangtze River Scientific Research Institute 27 (7), 1e5 (in Chinese with Englishabstract).

Milliman, J.D., Farnsworth, K.L., 2011. River Discharge to the Coastal OceandaGlobal Synthesis. Cambridge University Press, Cambridge, UK, p. 392.

Milliman, J.D., Shen, H.T., Yang, Z.S., Meade, R.H., 1985. Transport and deposition ofriver sediment in the Changjiang estuary and adjacent continental shelf. Con-tinental Shelf Research 4 (1/2), 37e45.

Phillips, J.D., 2003. Alluvial storage and the long-term stability of sediment yields.Basin Research 15, 153e163.

Phillips, J.D., Slattery, M.C., 2006. Sediment storage, sea level, and sediment deliveryto the ocean by coastal plain rivers. Progress in Physical Geography 30 (4),513e530.

Phillips, J.D., Slattery, M.C., Musselman, Z.A., 2004. Dam-to-delta sediment inputsand storage in the lower Trinity River, Texas. Geomorphology 62, 17e34.

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045

Z. Dai et al. / Quaternary International xxx (2015) 1e13 13

Pitt, J., 2001. Can we restore the Colorado River delta? Journal of Arid Environments49, 211e220.

Simmons, H.B., 1955. Some effects of upland discharge on estuarine hydraulics,1955. Separate Paper No. 792. In: Proceedings ASCE, vol. 81. ASCE Press, Reston,VA.

Spearman, J.R., Dearnale, M.P., Dennis, J.M., 1998. A simulation of estuary responseto training wall construction using a regime approach. Coastal Engineering 33,71e89.

Syvitski, J.P.M., V€or€osmarty, C.J., Kettner, A.J., Green, P., 2005. Impact of humans onthe flux of terrestrial sediment to the global coastal ocean. Science 308,376e380.

Syvitski, J.P.M., Kettner, A.J., Hannon, M.T., Hutton, E.W.H., Overeem, I.,Brakenridge, G.R., Day, J., V€or€osmarty, C., Saito, Y., Giosan, L., Nicholls, R.J., 2009.Sinking deltas. Nature Geoscience 2, 681e689.

Thomas, C.G., Spearman, J.R., Turnbull, M.J., 2002. Historical morphological changein the Mersey estuary. Continental Shelf Research 22, 1775e1794.

Van der Wal, D., Pye, K., Neal, A., 2002. Long-term morphological change in theRibble Estuary, northwest England. Marine Geology 189, 249e266.

Van der Wegen, M., Roelvink, J.A., 2008. Long-term morphodynamic evolution of atidal embayment using a two-dimensional, process-based model. Journal ofGeophysical Research 113, C03016. http://dx.doi.org/10.1029/2006JC003983.

Van Landeghem, K.J.J., Uehara, K., Wheeler, A.J., Mitchell, N.C., Scourse, J.D., 2009.Post-glacial sediment dynamics in the Irish Sea and sediment dynamics in theIrish Sea and sediment wave morphology: data-model comparisions. Conti-nental Shelf Research 29, 1723e1736.

Please cite this article in press as: Dai, Z., et al., Morphological evolution oQuaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.0

Wallace, J.W., Dickinson, 1972. Empirical orthogonal representation of time series inthe frequency domain, part 1: theoretical consideration. Journal of AppliedMeteorology 11, 887e892.

Walling, D.E., Fang, D., 2003. Recent trends in the suspended sediment loads of theworld rivers. Global and Planetary Change 39, 111e126.

Wang, J., Bai, S.B., Liu, P., Li, Y.Y., Gao, Z.G., Qu, G.X., Gao, G.G., 2009. Channel sedi-mentation and erosion of the Jiangsu reach of the Yangtze River during the last44 years. Earth Surface Processes and Landforms 34, 1587e1593.

Wang, Y.H., Dong, P., Oguchi, T., Chen, S.L., Shen, H.T., 2013. Long-term (1842e2006)morphological change and equilibrium state of the Changjiang (Yangtze) Es-tuary, China. Continental Shelf Research 56, 71e81.

Yang, S.L., He, S.L., Xie, W.H., 1998. The formation and evolution of the Jiudusha tidalisland as well as their relation to the development of the North and Southpassages in the Yangtze river estuary. The Ocean Engineering 16 (4), 55e64 (inChinese with English abstract).

Yang, S.L., Zhang, J., Xu, X.J., 2007. Influence of the Three Gorges Dam on downstreamdelivery of sediment and its environmental implications, Yangtze River. Geophys-ical Research Letters 34, L10401. http://dx.doi.org/10.1029/2007GL029472.

Yang, S.L., Milliman, J.D., Li, P., Xu, K., 2011. 50,000 dams later: erosion of the YangtzeRiver and its delta. Global and Planetary Change 75, 14e20.

Yun, C.X., 2004. Evolutionary Rules of the Changjiang Estuary in Recent Decades.China Ocean Press, Beijing (ISBN: 750276219).

Zhou, C.X., 1983. On Sandwave in Yawosha Channel of the Yangtze Estuary. In:Proceedings of the Second International Symposium on River Sedimentation.Water Resources and Electric Power Press, Beijing, pp. 650e661 (in Chinese).

f the South Passage in the Changjiang (Yangtze River) estuary, China,1.045